Alex Xi's Notes

Tag: moe

8 items with this tag.

  • Jul 06, 2026

    tweet-anti-moe-take

    • moe
    • architecture
    • contrarian
  • Jul 06, 2026

    tweet-deepseek-v4-analysis

    • deepseek
    • architecture
    • moe
    • scaling
  • Jul 06, 2026

    tweet-moe-training-instability

    • moe
    • training-stability
    • architecture
  • Jul 06, 2026

    mixture-experts-architectures

    • mixture-of-experts
    • moe
    • architecture
    • scaling
  • Jul 06, 2026

    prefill-as-a-service

    • inference
    • kv-cache
    • disaggregation
    • moe
    • infrastructure
  • Jul 06, 2026

    supporting-mixtral-gpt-fast

    • inference
    • moe
    • torch-compile
    • gpu-kernels
  • Jul 06, 2026

    cursor-warp-decode

    • inference
    • moe
    • gpu-kernels
    • blackwell
    • cursor
  • Jul 06, 2026

    frontier-pretraining-tpu-gpt-oss

    • pretraining
    • tpu
    • jax
    • infrastructure
    • moe
    • open-source